COIG-CQIA：Quality is All you need for Chinese Instruction Fine-tuning

Dataset Details

Dataset Description

欢迎来到COIG-CQIA，COIG-CQIA全称为Chinese Open Instruction Generalist - Quality is All You Need， 是一个开源的高质量指令微调数据集，旨在为中文NLP社区提供高质量且符合人类交互行为的指令微调数据。COIG-CQIA以中文互联网获取到的问答及文章作为原始数据，经过深度清洗、重构及人工审核构建而成。本项目受LIMA: Less Is More for Alignment等研究启发，使用少量高质量的数据即可让大语言模型学习到人类交互行为，因此在数据构建中我们十分注重数据的来源、质量与多样性，数据集详情请见数据介绍以及我们接下来的论文。

Welcome to the COIG-CQIA project page. COIG-CQIA stands for Chinese Open Instruction Generalist - Quality is All You Need, a high-quality Chinese instruction fine-tuning dataset. This dataset is designed to provide the Chinese NLP community with high-quality and human interaction-aligned instruction fine-tuning data.Inspired by studies like LIMA: Less Is More for Alignment, COIG-CQIA focuses on creating a dataset from Chinese internet sources including Q&A and articles. These are deeply cleansed, restructured, and manually reviewed to ensure quality, diversity, and relevance.

• Curated by: 来自零一万物、中科院深圳先进技术研究院，和M-A-P等机构的研究者们。
• Language(s) (NLP): 本数据集主要语言为中文。
• License: [More Information Needed]

本数据集当前为v0.1版本，如果您在使用中发现数据集存在问题或者有可以改进的地方，欢迎留言交流！

Uses

Direct Use

本数据集适用于指令微调，训练模型具备响应指令的能力。

Out-of-Scope Use

[More Information Needed]

数据

数据格式

{
    "instruction": "示例问题或者指令。",
    "input": "示例问题或指令的补充。",
    "output": "对输入的回复。",
    "task_type": {
        "major": ["问答"],
        "minor": ["百科问答"]
    },
    "domain": ["百科", "医疗"],
    "answer_from": "human",
    "human_verified": true,
    "copyright": "作者及版权信息。",
}

数据字段

• instruction: 用于输入的指令或者问题。
• input: 问题或指令的补充内容。
• output: 输入对应的回答。
• task_type: 表示该数据所属的主要任务类型和细分任务类型。
• domain: 该数据所属领域。
• answer_from: 回答是人类撰写的还是大模型撰写的，本数据集中绝大部分是由人类撰写的回答，少部分由大模型生成（经过了人工验证）。
• human_verified: 该数据是否有人类核验过。
• copyright: 包括该数据的版权信息，包括作者等。

当前版本的数据字段中仍有不完善的部分，我们将在近期的下一版本中补充。

数据详情

Recommendations

Users should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.

Citation

如果本项目为您的研究带来了帮助，欢迎引用！

@misc{bai2024coig,
    title={COIG-CQIA: Quality is All You Need for Chinese Instruction Fine-tuning}, 
    author={Bai, Yuelin and Du, Xinrun and Liang, Yiming and Jin, Yonggang and Liu, Ziqiang and Zhou, Junting and Zheng, Tianyu and Zhang, Xincheng and Ma, Nuo and Wang, Zekun and others},
    year={2024},
    eprint={2403.18058},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
  }

本数据集中也包含了以下公开数据：

@misc{zhang2023chinese,
    title={Chinese open instruction generalist: A preliminary release}, 
    author={Zhang, Ge and Shi, Yemin and Liu, Ruibo and Yuan, Ruibin and Li, Yizhi and Dong, Siwei and Shu, Yu and Li, Zhaoqun and Wang, Zekun and Lin, Chenghua and others},
    year={2023},
    eprint={2304.07987},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
  }
@misc{Firefly,
  author = {Jianxin Yang},
  title = {Firefly(流萤): 中文对话式大语言模型},
  year = {2023},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/yangjianxin1/Firefly}},
}
@misc{xu2023cvalues,
    title={CValues: Measuring the Values of Chinese Large Language Models from Safety to Responsibility}, 
    author={Guohai Xu and Jiayi Liu and Ming Yan and Haotian Xu and Jinghui Si and Zhuoran Zhou and Peng Yi and Xing Gao and Jitao Sang and Rong Zhang and Ji Zhang and Chao Peng and Fei Huang and Jingren Zhou},
    year={2023},
    eprint={2307.09705},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
  }